A5007620776- Natural Language Processing Techniques
- Topic Modeling
- Speech and dialogue systems
- Software Engineering Research
- Model-Driven Software Engineering Techniques
- Magnetic properties of thin films
- Security in Wireless Sensor Networks
- Cellular Automata and Applications
- Wikis in Education and Collaboration
- Advanced Manufacturing and Logistics Optimization
- Error Correcting Code Techniques
- Sentiment Analysis and Opinion Mining
- Advanced Data Compression Techniques
- Semantic Web and Ontologies
- Parallel Computing and Optimization Techniques
- Multimedia Communication and Technology
- Educational Robotics and Engineering
- Big Data and Business Intelligence
- Emotion and Mood Recognition
- Computational and Text Analysis Methods
- BIM and Construction Integration
- Medical Image Segmentation Techniques
- Web Data Mining and Analysis
- Speech Recognition and Synthesis
- Interconnection Networks and Systems
Dongguk University
1996-2023
University of Delaware
1991-1992
In this paper we provide an implementation strategy to map a functional specification of utterance into syntactically well‐formed sentence. We do by integrating the and syntactic perspectives on language, which take be exemplified systemic grammars tree adjoining (TAGs) respectively. From borrow use networks choices classify set possible constructions. The expressed in input are mapped our generator structure as defined TAG. argue that TAG structures can appropriate structural units...
It has been hypothesized that Tree Adjoining Grammar (TAG) is particularly well suited for sentence generation. unclear, however, how a generation system based on TAG should choose among the syntactic possibilities made available in grammar. In this paper we consider question of what needs to be done generate with TAGs and explain provides necessary features. This approach compared other TAG-based systems. Particular attention given Mumble-86 which, like our system, makes choice...
Describes a Korean morphological analyzer which can be used as part of language processor for speech translation system. We have modified the CYK algorithm so that we are able to analyze many phenomena occurring in spontaneous such ellipsis, shorter words, poor and mispronounced on, also constructed rule set with 112 connection rules seven kinds dictionaries, there total about 81,000 keywords. Currently, achieved success rate 93.0% text corpus dialogue hotel reservation domain.
This paper describes a Korean morphological analyzer which can be used as part of language processor for speech translation system.We h v e modied the CYK algorithm so that we are able to analyze many phenomena occurring in spontaneous such ellipsis, shorter words, poor and mispronounced words on.And also have constructed rule set with 112 connection rules seven kinds dictionaries, there totally about 81,000 keywords.Currently, w c hieved success rate 93.0% text corpus dialogue hotel...
자연어 처리 기술을 이용해 진술의 신빙성을 판별하는 소프트웨어에 대한 연구가 대두되고 있다. 이때 피해자의 진술에서 빈번히 생략되는 주어의 정확한 복원이 이루어진다면 컴퓨터의 문장 이해를 통해 진술신빙성 판단 성능을 높일 수 기존 한국어 무형대용어 해결 연구에서 주로 사용되었던 머신러닝 및 딥러닝 모델은 명사구에 초점을 맞춘 것으로, 일정 범위 내에 선행어가 존재하여야 생략어 가능하다는 등의 한계점이 성폭력 피해자 면담 조사 진술서의 특성상 서술어에 맞춘다면 효율적인 가능하다. 이에 기반해 본 연구에서는 구문 분석기와 규칙기반 모델을 생략된 가해자 또는 주어를 단위로 복원하는 방법을 채택했다. 구축한 모델로 실제 기록물에서 가해자의 주어가 2,396문장에 대해 주어 복원 실험을 진행하였다. 모델의 성능은 F1 스코어는 90.79%이며 낮지 않은 수준으로 판단한다. Research on software that determines the credibility of...